**** BEGIN LOGGING AT Mon Mar 07 02:59:58 2016 Mar 07 12:47:33 jkridner: when will the wednesday irc meetings begin officially? Mar 07 12:48:01 * bradfa has a conflict this week Mar 07 14:19:17 how can I get a mentor invitation, I already submit a idea Mar 07 15:09:26 http://pastebin.com/Qr8g602z Mar 07 15:09:35 no luck with BBBlfs Mar 07 15:12:06 did you install automake? Mar 07 15:12:41 carpediem_: Looks like you need to do sudo apt-get install build-essentials Mar 07 15:12:52 did all that Mar 07 15:13:04 my bad, sudo apt-get install build-essential Mar 07 15:13:21 what OS are you running? Mar 07 15:13:29 distro Mar 07 15:13:32 ok, chmod +x autogen.sh Mar 07 15:13:37 ubuntu 14.04 LTS Mar 07 15:13:58 and should you be running it as root? I'm not sure. Mar 07 15:14:26 running it as root is not recommended Mar 07 15:14:54 yep Mar 07 15:15:14 could you provide the output of the following commands? Mar 07 15:15:36 from within the soruce directory: ls -l Mar 07 15:16:45 ls -l `which sh` Mar 07 15:19:02 autoconf --version Mar 07 15:24:08 should I wait for any more commands? :D Mar 07 15:26:41 carpediem__: is your sh executable in /bin/? '$ which sh' Mar 07 15:31:59 nah Mar 07 15:32:57 oh, that permission denied trying to run it not as root... something weird is happening Mar 07 15:34:15 like weird permissions on /media/aditya/Studies/GSOC/Beaglebone/BBBlfs-master/Git\ Original/ or something Mar 07 15:39:31 pastebin.com/TxT13KEd Mar 07 15:40:53 hmm, those are weird permissions... Mar 07 15:41:44 running 'which sh' gives output as '/bin/sh' Mar 07 15:42:27 internet connection is acting pretty weirdly... :( Mar 07 15:43:17 ok, try running (from in the BBBlfs directory!): '$ find . -type f -exec chmod 644 {} \;' Mar 07 15:43:30 and: '$ find . -type d -exec chmod 755 {} \;' Mar 07 15:44:01 then do '$ chmod +x autogen.sh compile_tools.sh' Mar 07 15:44:12 that should at least get all those permissions where they should be... Mar 07 15:46:58 done Mar 07 15:50:06 any different? Mar 07 15:53:32 alexhiam: There? Mar 07 15:53:43 somewhat Mar 07 15:53:57 jkridner|work: BeagleLogic got featured at Hacker News! https://news.ycombinator.com/item?id=11232144 Mar 07 15:53:58 alexhiam: haha!! :-D Mar 07 15:54:23 alexhiam: I could compile the beaglebone blue kernel.. Mar 07 15:54:35 alexhiam: But I got only the uimage file.. Mar 07 15:54:43 with the imu driver? Mar 07 15:55:51 alexhiam: Yeah.. But I did not use menuconfig.. I created this beaglebone_blue_defconfig file.. Mar 07 15:56:08 on your github? Mar 07 15:57:00 alexhiam: https://github.com/kiran4399/linux/blob/bb_blue/arch/arm/configs/beaglebone_blue_defconfig Mar 07 15:57:23 that's fine, menuconfig is just a way to generate that Mar 07 15:57:53 alexhiam: https://github.com/kiran4399/linux/blob/bb_blue/arch/arm/configs/beaglebone_blue_defconfig#L1840 Mar 07 15:57:55 is that fine? Mar 07 15:58:38 looks right to me at first glance Mar 07 15:59:01 alexhiam: What is my next step? Mar 07 16:00:40 oh, actually probably better to build that driver as a module. That way the blue and black could use the same kernel Mar 07 16:02:11 alexhiam: so you want me to change y to m? Mar 07 16:02:18 right Mar 07 16:05:51 alexhiam: done. Mar 07 16:06:39 Hey ! I am trying to build the kernels with Rpmsg support, but the menuconfig shows that the driver code is not present. Mar 07 16:06:57 i.e. its like an empty submenu Mar 07 16:07:42 and I am able to see the code inside the drivers/rpmsg Mar 07 16:08:45 but still the RPmsg options is empty in the menuconfig Mar 07 16:09:06 ZeekHuge: maybe something wrong in the Kconfig file? Mar 07 16:10:21 oh, jkridner|work: you know if the beaglebone blue uses the MPU-9150 IMU like the robotics cape, which is now 'not recommended for new designs', or if it's got the MPU-9250? Mar 07 16:11:18 alexhiam, actually i was able to reach that option earlier. But now, I am doing it again, and the option is unavailable Mar 07 16:11:31 and i removed everything. Mar 07 16:11:39 started from scratch Mar 07 16:12:13 with a new unzipped source. but the option is unavailable Mar 07 16:13:56 unzipped source... so you have a clone of beagleboard/linux and you're adding the rpmsg source from a zip you downloaded from TI? Mar 07 16:15:49 No, i haven't cloned complete repo. Just downloaded the zip file of the current version of repo. That is the zip file i am talking about Mar 07 16:16:21 repo = beagleboard/linux Mar 07 16:16:41 oh, you should be working from a git clone for sure Mar 07 16:17:02 git is pretty integral to the Linux build system Mar 07 16:17:58 and using git is a requirement for beagleboard gsoc Mar 07 16:19:32 alexhiam: Is the ardupilot properly ported to bb black? Mar 07 16:19:51 alexhiam: I can lots of "To be done"... Mar 07 16:19:58 alexhiam: can see* Mar 07 16:20:00 define properly... Mar 07 16:20:16 I haven't actually looked at that port much, so I'm not the person to ask Mar 07 16:20:26 alexhiam: http://dev.ardupilot.com/wiki/building-for-beaglebone-black-on-linux/ Mar 07 16:20:51 alexhiam: what is the difference between beaglepilot and ardupilot in context to bb black? Mar 07 16:21:34 beaglepilot was the port that vmayoral did for the 2014 gsoc: http://elinux.org/BeagleBoard/GSoC/2014_Projects#Project:_BeaglePilot Mar 07 16:22:01 not sure if that was merged into ardupilot or what Mar 07 16:22:39 anyone invite vmayoral to mentor this year? Mar 07 16:26:42 alexhiam: I am not sure how.. but suddenly by bb black stoped working.. when I power it up.. the power led lights up for some time and goes off.. Mar 07 16:26:58 can you suggest me something.. Mar 07 16:41:27 kiran4399: hook up a serial cable to the bb and check the console output Mar 07 16:49:48 You mean that ftdi port? Mar 07 17:11:00 alexhiam: it is 9250 Mar 07 17:11:24 alexhiam: I've got someone asking me questions about fade.py not working on 4.1 kernels. Mar 07 17:12:07 yeah, PyBBIO is still not 4.1 compatible, haven't had the time :( Mar 07 17:12:24 * alexhiam needs more volunteers! Mar 07 17:14:16 kiran4399: yeah, that serial port is the place to start Mar 07 17:15:25 alexhiam: what you really need is minions Mar 07 17:16:14 minions would be acceptable Mar 07 18:28:26 yeah, take your eyes off the minions for a minute they'll rewrite your stack for rpi... Mar 07 18:29:15 either I have bad timing or students aren't coming in early .... Mar 07 19:12:29 * jkridner keeps getting damn direct e-mails. :( Mar 07 19:13:24 are they more specific then the few mails that got to the -gsoc list? :D Mar 07 19:15:31 jkridner: your email adress is on the org page: https://summerofcode.withgoogle.com/organizations/4817552005922816/ Mar 07 19:15:40 can you make that the ml instead? Mar 07 19:15:51 or remote that contact link Mar 07 19:16:03 remove* Mar 07 19:25:03 created gsoc @bb.org address Mar 07 19:25:16 should send to the list now. Mar 07 19:25:35 nice Mar 07 19:44:00 alexhiam, Just wanted to ask that I would have to compile the DT overlay I write?For that which DTC can be used?Like any DTC can be used or we would stick to some specific one? Mar 07 19:44:45 chanakya_vc: just use the version that ships with the debian image Mar 07 19:45:49 okay.But i believe the kernel of all debian based OS are all same?So does it matter? Mar 07 19:46:44 well, from http://beagleboard.org/latest-images the debian 7.9 image has 3.8 and the 8.3 image has 4.1, so use the 8.3 image Mar 07 19:47:17 Okay got it.Thanks Mar 07 19:51:17 carpediem_ whay are you trying to build BBBlfs as root? Mar 07 19:56:55 my guess it the folder you are using does not have the appropriate permissions to start Mar 07 19:59:02 where is the beaglepilot code? Mar 07 19:59:04 https://github.com/BeaglePilot/PPM-PRU Mar 07 19:59:12 https://github.com/BeaglePilot/PWM-PRU Mar 07 19:59:15 m_w: Ardupilot Mar 07 19:59:25 https://github.com/diydrones/ardupilot Mar 07 19:59:36 pretty well got merged. Mar 07 20:00:34 what is with all of the empty repos? Mar 07 20:01:57 I tried running the binary as well inside of /bin Mar 07 20:02:16 still gives the error : Command not found Mar 07 20:02:20 which binary Mar 07 20:02:21 ? Mar 07 20:02:28 inside of /bin Mar 07 20:02:36 ./flash_script.sh Mar 07 20:02:45 sorry, its not a binary Mar 07 20:02:51 but it should work Mar 07 20:03:01 let see Mar 07 20:03:08 use rcn ubuntu/debian and the kernel images are all there Mar 07 20:03:27 and I tried building it as root as well as using normal bash sudo commands Mar 07 20:03:30 do you have bash installed? Mar 07 20:03:31 multiple versions even Mar 07 20:03:35 ;) Mar 07 20:03:55 multiple versions of bash? No :D Mar 07 20:04:06 no, mutiple kernels Mar 07 20:04:12 No Mar 07 20:04:37 can you download/flash debian machinekit image? Mar 07 20:04:58 *flash as in let it install to emmc Mar 07 20:05:09 yeah. Mar 07 20:05:24 Did not flash to emmc, rather booted from it Mar 07 20:05:25 then you can boot different cards and still have current/stable debian+tools Mar 07 20:06:01 I will have to check, maybe I flashed Android to the board Mar 07 20:06:05 all of the current images should have robert's kernel/u-boot repos already Mar 07 20:07:21 otherwise go to https://eewiki.net/display/linuxonarm/BeagleBone+Black and pick your kernel and build it on your desktop Mar 07 20:07:34 jkridner: Did the PPM and PWM PRU ever get merged? Mar 07 20:08:28 or even coded? Mar 07 20:09:55 carpediem_: use am33x-v4.1 (Longterm 4.1.x) or the rt version Mar 07 20:10:53 also https://eewiki.net/display/linuxonarm/BeagleBone+Black#BeagleBoneBlack-capemgr:v4.1.x+ Mar 07 20:12:11 I know ds2 thinks it is a waste of the PRU but I am still interested Mar 07 20:12:55 m_w: https://github.com/diydrones/ardupilot/blob/master/libraries/AP_HAL_Linux/RCOutput_PRU.cpp Mar 07 20:14:37 /dev/mem ?!?! Mar 07 20:15:48 so this code controller 12 PWM outputs? Mar 07 20:15:54 controls Mar 07 20:16:19 m_w: I'm just googling code for you. :-) Mar 07 20:17:50 _av500_: have you had any of the IRC meetings? going to have one this Wednesday? Mar 07 20:18:21 jkridner: Thanks Mar 07 20:19:48 can we get bb.org mugs like these? http://dev.gentoo.org/~chithanh/raspberrypi/DSC00011.JPG Mar 07 20:19:58 minus the rpi3... Mar 07 20:37:20 I did some with Vistaprint. I could probably provide mug swag to mentors/students around the end. Mar 07 20:44:47 has anyone latched onto the serial terminal project yet? Mar 07 20:45:23 I wouldn't mind designing the PCB so that the students can focus on the software. Mar 07 20:47:37 also how is the testing of the beaglebone blue going? Mar 07 20:48:18 m_w: we have interest and there was at least one question on it Mar 07 20:48:59 plus that's the piece we need so have at it Mar 07 20:49:00 nerdboy: you think they wanted (or could) to do the hardware design? Mar 07 20:49:25 can we get both pins and db9s ? Mar 07 20:49:45 no, that's the piece we were a little leery of Mar 07 20:50:18 okay then lemme put it in my design queue Mar 07 20:50:20 student question didn't come up again that i saw so no details Mar 07 20:50:41 m_w: Blue is moving along. Never as fast as I'd want. Mar 07 20:50:59 nerdboy: ttl on the headers? Mar 07 20:51:30 yeah, that's correct for ftdi right? Mar 07 20:51:46 jkridner: any hardware bugs uncovered yet? Mar 07 20:51:49 like the current pins? Mar 07 20:52:14 which current pins? Mar 07 20:53:01 m_w: still doing cost reductions. Mar 07 20:53:03 the uart header on the bbb now Mar 07 20:53:55 so TTL 3.3V Mar 07 20:54:16 sounds right Mar 07 20:54:25 no problem Mar 07 20:54:50 do we want to make the TTL level configurable? Mar 07 20:55:11 that would help for 1.8V UARTs for instance Mar 07 20:55:52 sure Mar 07 20:56:28 how much handshaking for each port? Mar 07 20:56:38 if we had one now i would test in a customer box Mar 07 20:56:49 TX, RX Mar 07 20:56:54 RTS, CTS Mar 07 20:57:08 that can all be optional? Mar 07 20:57:31 sort of Mar 07 20:57:38 with db9s you'd want to be standard-ish Mar 07 20:57:58 the control would have to be there and just optionally used Mar 07 20:58:08 enough to connect with console ports/x86 serial stuff Mar 07 20:58:30 the pins could be more flexible Mar 07 20:58:39 console port can be just TX and RX Mar 07 20:58:46 * nerdboy was just looking at ping sensors Mar 07 20:59:14 at least one was like the simplest serial interface ever Mar 07 20:59:47 how about full handshaking? Mar 07 21:00:03 tx rx rts cts dsr dtr dcd Mar 07 21:00:12 i don't have a req for that but other people might Mar 07 21:00:27 rack of modems on bbb? Mar 07 21:00:34 a modem might need these Mar 07 21:00:39 yup Mar 07 21:01:14 how many ports we shooting for? Mar 07 21:01:26 all of them i thought... Mar 07 21:01:51 all of which ones? Mar 07 21:02:00 our current customer req is up to 5 Mar 07 21:02:43 so how about 8? Mar 07 21:02:49 there's one set of pins and 4 on the header i think Mar 07 21:03:09 oh, you mean via the i2c thing too Mar 07 21:03:17 yeah Mar 07 21:03:32 8 would be a nice/smallish number for a case Mar 07 21:03:55 16 then? :D Mar 07 21:04:37 in that case i would just get 2 Mar 07 21:05:01 I am confused Mar 07 21:05:14 if you can fit 3 on each side and 2 on the end or 4 per side the case could be small Mar 07 21:05:19 er Mar 07 21:05:39 now I understand Mar 07 21:05:46 the enclosure Mar 07 21:05:49 i just meant if i needed more than 8 serial ports i'd use 2 bbb Mar 07 21:06:03 keep is small and cheap Mar 07 21:06:38 we can keep the PCB smaller using 10pin to DB9 adapters Mar 07 21:07:04 we used those on all of our standard boards at EMAC inc Mar 07 21:07:18 with 8 ports the board could be bbb-size? Mar 07 21:07:26 http://ecx.images-amazon.com/images/I/61GH4YFHK-L._SY355_.jpg Mar 07 21:07:55 could be stackable i suppose Mar 07 21:08:07 perhaps Mar 07 21:08:27 the I2C would be overwhelmed after a while Mar 07 21:08:34 * nerdboy still has a bunch of those on brackets Mar 07 21:08:41 parallel too Mar 07 21:09:04 maybe 8 per i2c bus? Mar 07 21:09:36 I2C is typically 400Khz right? Mar 07 21:09:42 that would saturate i2c with all 8 going full speed i think Mar 07 21:09:56 it would puke Mar 07 21:10:14 SPI can going faster and probably handle more Mar 07 21:10:42 the chips were multi protocol Mar 07 21:10:46 yeah, i was just thinkin' the same thing... Mar 07 21:11:06 *spi has more bandwidth Mar 07 21:11:13 stacking doesn't work as well with SPI though Mar 07 21:11:14 not sure i'd want to do it Mar 07 21:11:29 why not? Mar 07 21:12:11 8-port console server would work, 8 port terminal server for ISP dialup probably not Mar 07 21:13:03 ah Mar 07 21:13:11 maybe 4 on uarts and 4 on i2c? Mar 07 21:14:20 maybe Mar 07 21:14:27 student should prototype and benchmark some i think Mar 07 21:14:50 https://www.maximintegrated.com/en/products/interface/controllers-expanders/MAX14830.html Mar 07 21:15:50 huh Mar 07 21:16:14 it is a quad i2c/spi uart Mar 07 21:16:59 yeah, i see that Mar 07 21:17:16 "huh" == "well i'll be a dirty..." Mar 07 21:17:40 what about peak throughput? Mar 07 21:17:53 * nerdboy didn't see that skimming datasheet Mar 07 21:21:41 it would get flustered at 115200 with I2C 400kHz mode if all of the port went simulataneously Mar 07 21:22:46 buffer overflows out the wazzo Mar 07 21:24:09 yeah, the applications listed don't include anything that looks high speed/low drag Mar 07 21:24:32 control stuff and messaging Mar 07 21:24:59 which is what we need anyway (for the mentioned application) Mar 07 21:25:47 * nerdboy wondering how the onboard uarts/arm core would do in that case... Mar 07 21:31:31 http://elinux.org/BeagleBone_Serial_CAN_RS485_RS232_Cape Mar 07 21:31:45 http://www.ebay.com/itm/like/261692397694?lpid=82&chn=ps&ul_noapp=true Mar 07 21:32:24 so, either one quad-chip on i2c and 4 on uarts or 2 quads, each on their own i2c bus? Mar 07 21:32:33 discuss... Mar 07 21:32:44 I can deal with either Mar 07 21:33:32 is there any reason not to use SPI? Mar 07 21:34:36 I spent a bunch of time getting omap2-mcpsi GPIO chipselects working Mar 07 21:36:21 not that we would need it for only 8 output ports if we use the quad serial chip Mar 07 21:37:13 <_av500_> jkridner: ack for irc meetink Mar 07 21:37:27 _av500_: cool Mar 07 21:37:41 <_av500_> jkridner: im kind of swamped in stuffs Mar 07 21:37:48 <_av500_> but I can do the meetings Mar 07 21:37:50 :( Mar 07 21:37:55 and the SPI would allow more throughput Mar 07 21:38:02 <_av500_> jkridner: can you do the invite Mar 07 21:38:13 getting some other admins engaged would really, really help. Mar 07 21:38:19 will try. Mar 07 21:39:05 <_av500_> I'll try to get a list of email ads and send you an email tomorrow Mar 07 21:39:58 http://lxr.free-electrons.com/source/drivers/tty/serial/max310x.c Mar 07 21:40:19 spi is supported in the mainline Mar 07 21:45:58 and we could get a leg up on programming if we order one of these: Mar 07 21:46:04 http://www.digikey.com/product-detail/en/maxim-integrated/MAX14830EVKIT%23/MAX14830EVKIT%23-ND/2812013 Mar 07 22:05:14 sounds good, jkridner is the one who buys stuff ;) Mar 07 22:05:40 jkridner: are the admin duties posted anywhere? Mar 07 22:07:10 nope.... responsibility is world domination. Mar 07 22:08:35 http://media-cache-ak0.pinimg.com/736x/36/2d/86/362d8673d368e0cf797b8fe3661b3da2.jpg Mar 07 22:19:31 hi, I'm doing the NOR Flash Emulator project, As the cape project (schematic and board) is not taken into consideration in GSOC, I'm doing it in my free time. Would you like to take a look on the scheme before routing? Mar 07 22:19:38 https://drive.google.com/file/d/0B1Sxahs4DCgrb0prWnpkSUswaEE/view Mar 07 22:20:14 sure Mar 07 22:22:30 m_w: you've outed yourself... Mar 07 22:22:48 how is that? Mar 07 22:24:11 pmezydlo: this schematic is very straightforward Mar 07 22:26:44 m_w: you are required only level voltage shifting, Mar 07 22:27:26 I think so Mar 07 22:27:37 lemme look at the proposal Mar 07 22:28:51 so it seems that you are to emulate a SPI flash Mar 07 22:29:14 and you want to be able to have board with different IO voltages connect Mar 07 22:30:17 what is the maximum SPI clock frequency you want to emulate? Mar 07 22:32:20 i think between 20-25Mhz Mar 07 22:32:36 max McSPI speed is 48Mhz Mar 07 22:32:45 okay then your design should be fine Mar 07 22:34:00 oh wait Mar 07 22:34:07 I see a bug Mar 07 22:34:35 the bypass caps are hooked up wrong Mar 07 22:38:19 m_w:i see, thanks Mar 07 22:39:11 send me a snapshot after the fix and with the values shown for the passives Mar 07 22:53:44 m_w: do you think about switch led by transistor? Mar 07 22:54:10 depends on the LED Mar 07 22:55:56 25mA Mar 07 22:56:11 20mA Mar 07 22:56:31 forward voltage? Mar 07 22:58:05 i think, around 2v Mar 07 22:58:09 as long as the forward voltage is less than 3.3V Mar 07 22:58:32 and the proper current limiting resistor is used you won't need a transistor Mar 07 23:00:31 470 ohms should be fine Mar 07 23:02:35 yes i calculated value is around 270, 470 is good Mar 07 23:03:19 470 is safer Mar 07 23:03:33 some of the GPIOs can only source 4mA Mar 07 23:04:14 https://learn.adafruit.com/blinking-an-led-with-beaglebone-black/overview Mar 07 23:07:16 new schematic and values are right side https://drive.google.com/open?id=0B1Sxahs4DCgrTDJzMFpaZlR5SkU Mar 07 23:07:41 this link is not public Mar 07 23:08:24 im sorry https://drive.google.com/file/d/0B1Sxahs4DCgrTDJzMFpaZlR5SkU/view?usp=sharing Mar 07 23:09:29 the bypass caps are still hooked up wrong Mar 07 23:11:01 VCC goes to the pin and one side of the cap Mar 07 23:11:11 GND goes to the other side of the cap Mar 07 23:13:18 and .1uF is probably better than 15pF Mar 07 23:14:31 already correcting Mar 07 23:14:39 http://www.electro-labs.com/bypass-capacitors-why-and-how-to-use-them/ Mar 07 23:14:56 I am ashamed Mar 07 23:15:09 it happens to the best Mar 07 23:15:21 not me though ;) Mar 07 23:29:03 i think schematic is finished https://drive.google.com/file/d/0B1Sxahs4DCgrUGM5TUtxaFBseUk/view?usp=sharing Mar 07 23:30:55 sure thats better Mar 07 23:31:23 I can look at the layout when it is ready Mar 07 23:31:53 Tomorrow, I will try to routing Mar 07 23:32:23 okay Mar 07 23:33:06 how is your soldering? Mar 07 23:33:21 these part are pretty small Mar 07 23:33:46 the ICs at least Mar 07 23:33:55 soldered already such Mar 07 23:34:17 how about 0603s? Mar 07 23:34:35 0805 Mar 07 23:35:25 do you think that they must be 0605? Mar 07 23:35:32 okay post a link and I will review. Mar 07 23:35:35 nah Mar 07 23:36:27 you have have plenty of board space :) Mar 07 23:39:00 m_w: Thanks for your reply. Good bye Mar 07 23:39:16 no problem **** ENDING LOGGING AT Tue Mar 08 02:59:59 2016